What's PREP's theory of Marital Failure?

[This reference work above is a wonderful resource packed with information on most of the major approaches to prevention and education efforts with couples.]

While there is not a singular pathway to marital distress or divorce, figure one portrays a comprehensive model that reflects a common path to marital failure based on findings from many different studies on marriage and relationships (from Stanley, 1997). In this model, two people initially become attracted to one another out of the mix of similarities and differences (and proximity). As time is spent together that is satisfying, a sense of attachment grows between the two partners. Along with this bond comes a sense of anxiety over the potential loss of the loved one. In this context, commitment develops, in large measure to remove this anxiety by the promise of a future together (Stanley, Lobitz, & Dickson, in press). For many couples, this commitment culminates in marriage.

Prior to the wedding day, most first time married couples have had few tests of their ability to handle conflict. They simply have not encountered many significant issues or disagreements during courtship. That is partly why satisfaction tends to be very high at this stage (Stanley & Markman, 1992). Yet, there's clear evidence that how couples communicate and handle conflict foretells an important story about their future--more important than their premarital level of satisfaction (Markman, 1981; Markman & Hahlweg, 1993). Over time, this committed couple must increasingly deal with the problems of life, together. This explains why so many couples can start out so committed and so happy only to find their attachment being eroded by the constant dripping of unresolved and upsetting conflicts.

What they argue about and how they argue is a function of both expectations and their abilities to communicate and negotiate effectively--most of which is based on previous experiences in life with family of origin, past relationships, and the cultural context. Certain patterns of mismanaged conflict that are destructive for relationships will be repeatedly expressed in many couples (Gottman, 1993; Gottman & Krokoff, 1989; Markman, Stanley, & Blumberg, 1994; Notarius & Markman, 1993). Out of this mix, a very important change occurs over time in the relationship: the presence of the partner becomes increasingly associated with pain and frustration, not pleasure or support. For most couples, this violates a basic assumption about what being together is about--having a most intimate and supportive friend for life.

Negative interpretations about the partner can become commonplace as a "me versus you" environment takes hold. These negative interpretations lead each to consistently interpret the actions of the other as more negative than is warranted (Baucom & Epstein, 1990). Confidence erodes and the sense of attachment that led to commitment in the first place becomes more fragile. In fact, we have observed that many couples will begin regularly threatening the future of the relationship as commitment becomes a pawn to be sacrificed in the heat of tormenting conflicts (Stanley, Lobitz, & Dickson, in press).

Ultimately, the marriage becomes a shell of its former state, with the view each partner holds of the other radically changing, as in the classic reversal of figure and ground (Gottman, 1993). In this case, the initial view becomes harder and harder to reinstate. At this point, the keys to the stability of the marriage lie less in commitment as dedication and more in commitment as constraint (Stanley & Markman, 1992). In other words, decisions to stay or leave now have more to do with the costs of leaving than the desire to stay. As constraints keeping marriages together weaken in our society, couples who get to this point are increasingly likely to divorce, whereas in the past many stayed together in stable, miserable marriages. Of course some couples rebound from this place of high constraint and low dedication to re-develop dedication, love, and positivity in their marriages (the third circle).

This model of marital failure suggests a number of targets for intervention. These include certain patterns of communication, conflict management, dysfunctional beliefs, and understandings and motivations regarding commitment. The good news is that couples are not predestined to this path of destruction. There are things they can learn and act on to stay off the path of failure and on the path of friendship, support, and intimacy. But many of these skills and attitudes are anything but natural for most couples.

Outcome Research
Excerpted review from:
Stanley, S.M., Markman, H.J., Prado, L.M., Olmos?Gallo, P.A., Tonelli, L., St. Peters, M., Leber, B.D., Bobulinski, M., Cordova, A., & Whitton, S. (Under review). Short term effects of premarital training in a religious, community based sample.

Studies on versions of PREP have been conducted by various researchers in a number of countries with varying results. Depending on the view taken, one can take the results to date as either greatly encouraging of prevention efforts or as examples of the ways in which the potential for selection effects can cloud interpretation of effects in outcome studies (Stanley & Markman, 1998). Despite the difficulties, outcome studies based on random trials provide excellent opportunities for testing the theories that underlie interventions (Coie et al., 1993).

In one of the most extensive, long-term studies on PREP, couples who took the program (the version existing at that time) before marriage had less negative interaction, more positive interaction, lower rates of relationship aggression, lower combined rates of breakup or divorce, and higher levels of relationship satisfaction up to five years following the training (e.g., Markman, Floyd, Stanley, & Storaasli, 1988; Markman, Renick, Floyd, Stanley, & Clements, 1993). However, due to differential attrition (more couples in the control group drop out of the study) and the probable wearing off of the effect over years of time, differences between controls and PREP couples after five years are equivocal. About one half to two thirds of the sample were randomly chosen to be offered the intervention, and only about one half of those who were offered it chose to take it. Thus, findings from this study are open to various interpretation because of potential selection effects, though we never found any systematic differences between acceptors and decliners at pre-test on a host of measures.

A long term study of a German adaptation of PREP was undertaken in Germany (Thurmaier, Engl, Eckert, & Hahlweg, 1993). At the five year point in this study on this German version of PREP, couples taking PREP premaritally had a divorce rate of 3% compared to 16% for control couples (Thurmaier, Engl, & Hahlweg, 1999). Furthermore, the couples taking PREP showed consistently more positive and less negative interaction at every assessment point after the training. The control group in this study was a mixture of no treatment and traditional premarital training as offered in the Catholic church in Germany. Couples were not randomly assigned to groups in this study, though the control group does have high external validity in that it reflects what couples often do for premarital training in Germany. Furthermore, the communication effects from the program (e.g., pre vs. post assessment) seem quite clear regardless of issues related to the nature of the control group. This same data set was subsequently analyzed using growth curve modeling by Bradbury, Thurmaier, Engl, Karney, Hahlweg, & Markman (1998). By analysis of intercepts, they found that couples who took PREP communicated better than control couples. More importantly, PREP couples displayed slower increases in negative verbal behavior, and wives who took this version of PREP showed slower declines in satisfaction over time.

Another premarital study on PREP compared the program with a version of Engaged Encounter, wherein Engaged Encounter was adapted to match PREP for time length and format (five, 2.5 hour sessions; Renick, Blumberg, & Markman, 1992). This study lacks long term follow up, however like the two just mentioned, it did employ objective coding of couple interaction. Further, this study used random assignment of couples into one of the two well regarded programs. At post test, the PREP couples had significantly improved on both objectively coded interaction and self-reported satisfaction in comparison to those receiving Engaged Encounter.

Another study on a variation of a PREP intervention (by Halford and colleagues in Australia) examined the effectiveness of such a preventive program with high and low risk couples in comparison to an alternate biblio-educational treatment. In this research, high risk was defined as parental divorce in the woman's family of origin or interparental aggression in the man's family of origin. Divorce of the female's parents, but not the male's, has been shown to increase a couple's premarital risk of negative communication (Sanders, Halford, & Behrens, 1999). At the one year point, the results demonstrated that the couples who went through this variation of PREP had good retention of the skills taught. At the four year follow up, low risk couples who took the intervention were doing no better than low risk couples who took the alternate intervention, with couples in both groups maintaining initial levels of satisfaction. However, high risk couples who took the intervention showed significant advantages in maintaining satisfaction compared to the those taking the alternative, with the high risk couples taking the alternate intervention deteriorating in satisfaction over the four year period (Behrens & Halford, 1994).

In a study of an adaptation of PREP for Holland, Van Widenfelt, Hosman, Schaap, and van der Staak (1996), using a higher risk group of premarital couples, found no differences between those taking the program and control couples. Like the 1980s Denver study, this study used random assignment but couples could self-select regarding participation. The results were particularly hard to interpret because of several methodological factors. First, there was differential attrition between the PREP and control groups, with the more distressed control couples being more likely to drop out of the study. This effect seems common to outcome studies on PREP (perhaps reflecting some treatment effect). Unfortunately, it makes interpretation of longer term results difficult since the control groups may become increasingly select for couples doing well over time, with potential differences between the groups becoming harder to measure. A second problem with the Holland research was that, following self-selection, the PREP couples had been together significantly longer than the control couples. Third, the PREP couples had been together an average of nine years prior to intervention, diminishing generalizations about prevention. Finally, the researchers in the Holland study did not report analyses on objective coding of the interaction data. In general, some of the most impressive findings in the studies on PREP-notwithstanding other methodological concerns-have come from designs utilizing objective coding of couple interaction behavior.

An unpublished study by Trathen (1995) compared a Christian version of PREP with a less interaction oriented intervention, but one that was likewise relatively conceptually full and lengthy (twelve hours like PREP). This study likewise did not show specific benefits favoring PREP. Both groups in this study were comprised of relatively religious subjects. In both groups, scores on commitment and confidence in the future of the relationship increased following training. As with the study in Holland, a tendency for differential drop out of subjects (more from the control group) and the lack of objective assessment of couple interaction make interpretation of the findings difficult. Nevertheless, the results suggest that other well conceived and delivered programs may yield positive results for couples. In part, this may be because any well conceived and presented program has the added benefit of causing couples both to wait longer to be married and to be more reflective about the decision they are making.

Family Stability Project
The Family Stability Project is our latest, large scale research effort looking at the effectiveness of PREP used as a premarital intervention. In the most recent funding period, we began the next stage of our program of prevention research and dissemination. Whereas the term "Stage II" has been given to tightly controlled clinical trials of efficacy, the term "Stage III" has been given to tests of the effectiveness of dissemination (NIMH, 1999). Using this terminology, the major overarching aim of this project is to conduct a Stage III test of a premarital divorce and marital distress prevention program disseminated in a community sample of religious organizations. As such, a major aim has been to evaluate the extent to which we could train clergy and lay leaders to effectively deliver the PREP program in religious settings. To this end, this project represents applied research rather than a one designed to determine the absolute highest efficacy level obtainable in a tightly controlled university laboratory environment. We are trying to determine how this intervention can best be implemented, and eventually contribute to the reduction of mental disorders.

We have established a study where we have recruited to date 109 religious organizations (ROs; from a vast variety of religious groups) and randomly assigned the ROs to one of three groups: 1) PREP delivered by trained clergy (ROPREP); 2) PREP delivered by our University of Denver team (DUPREP ); and 3) Naturally occurring premarital intervention services (NO), wherein the ROs employ whatever premarital services they normally use with their couples. To date, 64 of these 109 ROs have contributed couples to the study, resulting in 264 couples in the study at this time.

As we expected, the couples receiving the empirically based PREP intervention show strong, early advantages relative to those couples receiving more traditional premarital services in terms of interaction quality. By post assessment following premarital training, the couples taking PREP are communicating more positively and less negatively as rated by objective coders in our laboratory. Furthermore, the couples taking the mixture of naturally occurring premarital programs offered in ROs were communicating significantly more negatively and less positively by post assessment, compared to the couples taking PREP. Further, on short term results, there is no relationship between the amount of hours religious leaders spend in the naturally occurring programs with the couples and any of the outcome measures of couple functioning.

Importantly, these short term results favoring PREP are regardless of whether or not the couples were trained by our staff or their own clergy and lay leaders. This suggests that clergy and lay leaders can be trained to effectively present this program in a three day workshop. It is also clear that the couples taking PREP in either track were significantly more satisfied with their premarital education than couples receiving the more traditional offerings. This is reflected not only in the raw satisfaction data, but also in the fact that those couples were more likely to perceive that their leaders were working with goals and staying on tasks that were meaningful and logical to them. Not only were the couples more satisfied with the PREP intervention, clergy and lay leaders have demonstrated great interest in the ongoing use of the materials as part of their premarital and marital education regimens. In fact, as of the time of this writing, 22 of these leaders have already given PREP or parts of it to 310 other couples who are not in the study. Most importantly given the dissemination aims of this research, the short-term evidence suggests that the intervention is transportable to the community where religious leaders were able to achieve excellent short term effects on couple interaction.

Each subject was asked by open ended question what features of their premarital training they found most helpful. They could list up to three features, and answers were coded into categories. Given that the PREP tracks had training that was both uniform and substantially different from the NO track, the relative ranking of the ratings were very different depending on what kind of training subjects received. The rankings for both PREP tracks were essentially the same, so they are presented together here. The percentages reported here are the percentages of those subjects who mentioned the feature listed.

Males taking PREP reported that the following four features were most helpful: communication skills training (89%), format features (19%), communication lectures (15%), and expectations clarification (14%). Females taking PREP reported that the following four features were most helpful: communication skills training (90%), format features (22%), communication lectures (14%), and expectations clarification (21%). With regard to subjects in the PREP tracks ranking communication skills as the most helpful feature of their training, the "Speaker/Listener Technique" was specifically listed by many subjects as the most helpful feature of their training (males = 78%, females = 75%).

Males in the naturally occurring track reported that the following three features were most helpful: communication lectures (35%), insight oriented counseling (24%), and structured assessment (12%). Females in the naturally occurring track reported these as most helpful: communication lectures (39%), insight oriented counseling (33%), with a tie for third place between five other features (all rated at 11%).

In summary of the FSP, the initial results from this research are very consistent with our theory of how successful preventive interventions can modify patterns associated with risk early in relationships that may then translate into reduced risk on many other dimensions over time. At least with respect to pre-post, the intervention appears to be transportable to these community based providers, and that they appear to be as effective as our university staff in producing the hypothesized pre-post changes in objectively rated interaction. However, we caution that these results are based analysis of short-term data. Longer term follow up is needed to assess if and how the early differences in interaction quality affect other outcomes, and if the ROPREP results are equal to the DUPREP results, and superior to the NO results, over time.

Survey Research
Most preparation models are likely to be well received by couples as long as the models are well conceived, theoretically plausible, and thoughtfully presented. But some material will be perceived by the couples as more relevant than other material. In an important, large sample survey recently conducted in the Catholic Church, couples were asked many questions about their premarital preparation experiences. When asked what content areas were most helpful, the top three rankings went to the three "C's" of communication (73.5% rated as helpful), commitment (70.4% rated as helpful), and conflict resolution (67.2% rated as helpful; Center for Marriage and Family, 1995). Rated significantly lower were topics such as personality issues, finances, sex and intimacy, and career issues (see Silliman & Schumm, 1989, for similar findings). While it is doubtful that most of the premarital programs experienced by these couples were skills-based, it is very clear that these couples who were now well into their marriages recognized the central need for these foci in premarital training. Skills-based models almost always focus on communication. More specifically, PREP (and Christian PREP) focus on all three of these "C's" as domains for cognitive and behavioral change. (From Stanley, 1997)

The findings from the FSP mentioned above are entirely consistent with those noted above. Couples seem to see work on communication, conflict, and commitment as highly relevant in their desire to form and keep long lasting, happy marriages. These are the core themes of PREP.

Commitment Research
As a result of experiences with the longitudinal study begun in 1980, Stanley began exploring commitment as a major research focus (Stanley, 1986). Since 1984, Stanley has been developing various measures of commitment related dimensions to aid in better understanding what helps healthy and satisfied couples stay that way. Dr. Markman, Lisa LaViolette, and Elizabeth Collins have played roles in this work. Historically, there has been little research on commitment, especially given the obvious importance of the construct in marriage. However, there is an increasingly substantial body of theory and research related to commitment. The current versions of PREP as well as Christian PREP take advantage of this body of knowledge to help couples develop and strengthen their levels of mutual dedication to one another. This knowledge adds some cognitive oriented strategies to PREP that couples can employ in building strong marriages (Stanley & Markman, 1992; Stanley, Lobitz, & Dickson, 1999; Stanley, 1998).